On Mon, Mar 24, 2008 at 2:16 AM, Todd Walton <[EMAIL PROTECTED]> wrote:
> On Mon, Mar 24, 2008 at 3:14 AM, Bob La Quey <[EMAIL PROTECTED]> wrote:
> > I challenge each of you to state the problem tha you think
> > general purpose computing (or some subset, if you think
> > the challenge unfair) is trying to solve.
>
> Using a machine to do some complex task that we think it could do
> faster or more accurately than we can. "Complex" and "more
> accurately" being the subject of computing research.
>
> If you're going to ask such a fundamental question, you might take a
> step back and ask what "computing" is anyway. Does a hammer computer?
> Does a loom? Does computing mean numbers? The answers to these
> questions will define the limits of computing, and we could then be
> better prepared to ask what it is we think that "computing" will do
> for us.
>
> -todd
I guess Todd is the only wise man left :)
OK. I will toss something out. I do _not_ make
any pretension to deep thought here. Rather this
answer is simply intended as a seed for discussion.
General purpose computing solves the following problem:
Provide a means for a human to communicate
a problem description to a machine so that
the machine may in some way help solve the
problem.
Thus the focus is on problem description. A computing
language must, first and foremost, be a language for
describing problems.
If we go this route then we can begin by making a
list of problems, and abstracting them. Is this
what computer "science" is supposed to be about?
If not then what?
BobLQ
PS. Some might argue that language is only one of
several ways of describing a problem ... but for
me, for now, language will do.
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
